Automatic sub-cluster selection algorithm for the hierarchical clustering of file objects

A non-transitory storage medium comprising instructions that, when executed by one or more processors, cause the one or more processors to:

• access a dendrogram that includes a set of nodes comprising a parent node and one or more child nodes below the parent node, wherein each node in the dendrogram includes a respective distance measure in a descending order from the parent node to the one or more child nodes;
• beginning at the parent node of the dendrogram, such that a top-down approach is used:
• (i) calculate a first W-index for the parent node and a left child node, wherein the first W-index is a first ratio obtained by dividing a distance measure of the parent node by a distance measure of the left child node;
• and
• (ii) calculate a second W-index for the parent node and a right child node, wherein the second W-index is a second ratio obtained by dividing the distance measure of the parent node by a distance measure of the right child node;
• recursively calculate, via steps (i) and (ii) above, respective W-indices for all other parent nodes in the dendrogram;
• and
• use a maximum gap selection algorithm to form a plurality of clusters of nodes, wherein the maximum gap selection algorithm forms the plurality of clusters of nodes by cutting the dendrogram based on the W-indices, wherein said cutting results in formation of the clusters of nodes.